はてなブックマーク

A Semi-Naive Bayesian Classifier for Fast Patch Classification We show that it is possible to train a Semi-Naive Bayesian Classifier to recognize keypoints without the need for intensive preprocessing stages like fine-scale selection, dominant orientation and affine parameter estimation. We also demonstrate that this approach is superior to previous methods that use weighted averaging of Randomize

We show that it is possible to estimate depth from two wide baseline images using a dense descriptor. Our local descriptor, called DAISY, is very fast and efficient to compute. It depends on histograms of gradients like SIFT and GLOH but uses a Gaussian weighting and circularly symmetrical kernel. This gives us our speed and efficiency for dense computations. We compute 200-length descriptors for

PhD Thesis V. Lepetit. Gestion des occultations en Réalité Augmentée. Thèse d'université, Université Henri Poincaré, Nancy, May 2001. Book Chapter P. Fua, and V. Lepetit, Vision Based 3D Tracking and Pose Estimation for Mixed Reality, pp. 1-22 in M. Haller, M. Billinghurst, and B. Thomas (Eds.) Emerging Technologies of Augmented Reality, Interfaces and Design. Idea Group Publishing, 2007. Journal

BazAR is a computer vision library based on feature points detection and matching. In particular, it is able to quickly detect and register known planar objects in images. Well adapted to Augmented Reality applications, it is the result of advanced computer vision research. Watch some calibration screen shots or augmented reality videos. Downloads Source code is available under the GNU General Pub